Model Square
DeepSeek

deepseek-v4-pro

DeepSeek·Dynamic billing

Open MoE flagship with million-token context for coding and long agent runs

TPS
Average latency
Success rate

Pricing

Base Price

Input
$1.3235/ 1M
Output
$3.9706/ 1M
Cache Read$0.0441/ 1M
Dynamic billing
Prices vary by usage tier and request conditions
Tiered price table
peak
Input
$1.3235 / 1M
Output
$3.9706 / 1M
Cache Read
$0.0441 / 1M
off_peak
Input
$0.6618 / 1M
Output
$1.9853 / 1M
Cache Read
$0.0221 / 1M

Pricing by Group

default1x
TierInputOutputCache Read
peak
$1.3235
$3.9706
$0.0441
off_peak
$0.6618
$1.9853
$0.0221

Prices shown per 1M tokens

Model

ProviderDeepSeek
TypeDynamic billing
Groups
default
Endpoints
anthropicgeminiopenaiopenai_response
Tags
ReasoningToolsOpenWeights1M
deepseek-v4-pro